On the asymptotic properties of a feasible estimator of the continuous time long memory parameter

This article considers a fractional noise model in continuous time and examines the asymptotic properties of a feasible frequency domain maximum likelihood estimator of the long memory parameter. The feasible estimator is one that maximizes an approximation to the likelihood function (the approximation arises from the fact that the spectral density function involves the finite truncation of an infinite summation). It is of interest therefore to explore the conditions required of this approximation to ensure the consistency and asymptotic normality of this estimator.     

Testing for a break in persistence under long-range dependencies

Abstract.  We show that tests for a break in the persistence of a time series in the classical I (0)/ I (1) framework have serious size distortions when the actual data-generating process (DGP) exhibits long-range dependencies. We prove that the limiting distribution of a CUSUM of squares-based test depends on the true memory parameter if the DGP exhibits long memory. We propose adjusted critical values for the test and give finite sample response curves that allow easy implementation of the test by the practitioner and also ease in computing the relevant critical values. We furthermore prove the consistency of the test for a simple breakpoint estimator also under long memory. We show that the test has satisfying power properties when the correct critical values are used.     

Local Whittle estimation of the memory parameter in presence of deterministic components

We discuss the estimation of the order of integration of a fractional process that may be contaminated by a time‐varying deterministic trend or by a break in the mean. We show that in some cases the estimate may still be consistent and asymptotically normally distributed even when the order of magnitude of the spectral density of the fractional process does not dominate the one of the periodogram of the contaminating term. If trimming is introduced, stronger deterministic components may be neglected. The performance of the estimate in small samples is studied in a Monte Carlo experiment.     

Estimation of the long‐memory stochastic volatility model parameters that is robust to level shifts and deterministic trends

I provide conditions under which the trimmed FDQML estimator, advanced by McCloskey (2010) in the context of fully parametric short‐memory models, can be used to estimate the long‐memory stochastic volatility model parameters in the presence of additive low‐frequency contamination in log‐squared returns. The types of low‐frequency contamination covered include level shifts as well as deterministic trends. I establish consistency and asymptotic normality in the presence or absence of such low‐frequency contamination under certain conditions on the growth rate of the trimming parameter. I also provide theoretical guidance on the choice of trimming parameter by heuristically obtaining its asymptotic MSE‐optimal rate under certain types of low‐frequency contamination. A simulation study examines the finite sample properties of the robust estimator, showing substantial gains from its use in the presence of level shifts. The finite sample analysis also explores how different levels of trimming affect the parameter estimates in the presence and absence of low‐frequency contamination and long‐memory.     

Cocoa-butter long spacings and the memory effect

Depending on the maximum temperature before cooling, cocoa butter recrystallizes in the β(V) and/or β(VI) phase. In order to obtain structural information on the seed material initiating this recrystallization process, experiments with cocoa butter were performed at a small-angle X-ray scattering station. Crystallization of cocoa butter at various crystallization temperatures provided long d-spacing values (d>45 Å) that can be attributed to the polymorphic phases. From the recrystallization experiments it is concluded that the seeds initiating rapidstarting recrystallization have an 1,3-distearoyl-2-oleoylglycerol-dominated triple chain-length packing. Furthermore, the β(VI) phase of cocoa butter seems to adopt a similar packing. The seed crystals that initiate the slow-starting recrystallization and result in the β(V) phase are likely to be different from those giving the β(VI) phase.     

Temporal Aggregation and Bandwidth selection in estimating long memory

This article aims at showing that a temporal aggregation and a specific bandwidth reduction lead to the same asymptotic properties in estimating long memory by Geweke and Porter‐Hudak's [Journal of Time Series Analysis (1983 ) vol. 4, pp. 221–237] and Robinson's [Annals of Statistics (1995b ) vol. 23, pp. 1630–1661] estimators. In other words, irrespective of the level of temporal aggregation, the asymptotic properties of the estimator are uniquely determined by the number of periodogram ordinates used in the estimation, provided some mild additional assumptions are imposed. Monte Carlo simulations show that this result is a good approximation in finite samples. A real example with the daily US Dollar/French Franc exchange rate series is also provided.     

Likelihood inference for discriminating between long‐memory and change‐point models

We develop a likelihood ratio (LR) test procedure for discriminating between a short‐memory time series with a change‐point (CP) and a long‐memory (LM) time series. Under the null hypothesis, the time series consists of two segments of short‐memory time series with different means and possibly different covariance functions. The location of the shift in the mean is unknown. Under the alternative, the time series has no shift in mean but rather is LM. The LR statistic is defined as the normalized log‐ratio of the Whittle likelihood between the CP model and the LM model, which is asymptotically normally distributed under the null. The LR test provides a parametric alternative to the CUSUM test proposed by Berkes et al. (2006) . Moreover, the LR test is more general than the CUSUM test in the sense that it is applicable to changes in other marginal or dependence features other than a change‐in‐mean. We show its good performance in simulations and apply it to two data examples.     

Wavelet change‐point estimation for long memory non‐parametric random design models

For a random design regression model with long memory design and long memory errors, we consider the problem of detecting a change point for sharp cusp or jump discontinuity in the regression function. Using the wavelet methods, we obtain estimators for the change point, the jump size and the regression function. The strong consistencies of these estimators are given in terms of convergence rates.     

Fractional integration and structural breaks at unknown periods of time

Abstract. This article deals with the analysis of structural breaks in the context of fractionally integrated models. We assume that the break dates are unknown and that the different sub‐samples possess different intercepts, slope coefficients and fractional orders of integration. The procedure is based on linear regression models using a grid of values for the fractional differencing parameters and least squares estimation. Several Monte Carlo experiments conducted across the study show that the procedure performs well if the sample size is large enough. Two empirical applications are described at the end of the article.     

Robust discrimination between long‐range dependence and a change in mean

In this article we introduce a robust to outliers Wilcoxon change‐point testing procedure, for distinguishing between short‐range dependent time series with a change in mean at unknown time and stationary long‐range dependent time series. We establish the asymptotic distribution of the test statistic under the null hypothesis for L₁ near epoch dependent processes and show its consistency under the alternative. The Wilcoxon‐type testing procedure similarly as the CUSUM‐type testing procedure (of Berkes I., Horváth L., Kokoszka P. and Shao Q. 2006. Ann.Statist. 34:1140–1165), requires estimation of the location of a possible change‐point, and then using pre‐ and post‐break subsamples to discriminate between short and long‐range dependence. A simulation study examines the empirical size and power of the Wilcoxon‐type testing procedure in standard cases and with disturbances by outliers. It shows that in standard cases the Wilcoxon‐type testing procedure behaves equally well as the CUSUM‐type testing procedure but outperforms it in presence of outliers. We also apply both testing procedure to hydrologic data.     

Spurious regressions between I(1) processes with long memory errors

In this paper we develop the asymptotic distribution theory for spurious regression between I(1) processes with long-memory stationary errors. Our result departs from the standard results of Phillips (Understanding spurious regression in econometrics. J. Economet. 33 (1986), 311–40) in two respects. First, the limit theory we apply is based on a functional central limit theorem for stationary linear processes whose spectral density at frequency zero may diverge or collapse to zero. Second, different limit distributions may apply depending on the form of long memory exhibited by the error term. We also discuss the extension of our analyis to spurious regression with fitted intercept.     

Deep learning for quality prediction of nonlinear dynamic processes with variable attention-based long short-term memory network

Industrial processes are often characterized with high nonlinearities and dynamics. For soft sensor modelling, it is important to model the nonlinear and dynamic relationship between input and output data. Thus, long short-term memory (LSTM) networks are suitable for quality prediction of soft sensor modelling. However, they do not consider the relevance of different input variables with the quality variable. To address this issue, a variable attention-based long short-term memory (VA-LSTM) network is proposed for soft sensing in this paper. In VA-LSTM, variable attention is designed to identify important input variables according to their relevance with quality prediction. After that, different attention weights are calculated and assigned to further obtain a weighted input sample at each time step. Finally, the LSTM network is exploited to capture the long-term dependencies of the weighted input time series to predict the quality variable. The performance of the proposed modelling method is validated on an industrial debutanizer column and a hydrocracking process.     

Conditional variance estimation in regression models with long memory

In this article we study asymptotic properties of a non‐parametric kernel estimator of the conditional variance in a random design model with parametric mean and heteroscedastic errors, for a class of long‐memory errors and predictors. We establish small and large bandwidths asymptotics, which show a different behaviour compared with that of kernel estimators of the conditional mean. We distinguish between an oracle case (i.e. where the errors are directly observed) and a non‐oracle case (where the errors are replaced with residuals) and show non‐equivalence between the oracle and non‐oracle case. We also discuss a practical problem of bandwidth choice. Theoretical results are justified by simulation studies. We apply our theory to DJA and FTSE indices.     

Simulating a class of stationary Gaussian processes using the Davies–Harte algorithm, with application to long memory processes

We demonstrate that the fast and exact Davies–Harte algorithm is valid for simulating a certain class of stationary Gaussian processes – those with a negative autocovariance sequence for all non-zero lags. The result applies to well known classes of long memory processes: Gaussian fractionally differenced (FD) processes, fractional Gaussian noise (fGn) and the nonstationary fractional Brownian Motion (fBm).     

MODELLING FOR THE WAVELET COEFFICIENTS OF ARFIMA PROCESSES

We consider a model for the discrete nonboundary wavelet coefficients of autoregressive fractionally integrated moving average (ARFIMA) processes in each scale. Because the utility of the wavelet transform for the long‐range dependent processes, which many authors have explained in semi‐parametrical literature, is approximating the transformed processes to white noise processes in each scale, there have been few studies in a parametric setting. In this article, we propose the model from the forms of the (generalized) spectral density functions (SDFs) of these coefficients. Since the discrete wavelet transform has the property of downsampling, we cannot directly represent these (generalized) SDFs. To overcome this problem, we define the discrete non‐decimated nonboundary wavelet coefficients and compute their (generalized) SDFs. Using these functions and restricting the wavelet filters to the Daubechies wavelets and least asymmetric filters, we make the (generalized) SDFs of the discrete nonboundary wavelet coefficients of ARFIMA processes in each scale clear. Additionally, we propose a model for the discrete nonboundary scaling coefficients in each scale.     

Improved Tests for Forecast Comparisons in the Presence of Instabilities

Of interest is comparing the out‐of‐sample forecasting performance of two competing models in the presence of possible instabilities. To that effect, we suggest using simple structural change tests, sup‐Wald and UDmax for changes in the mean of the loss differences. It is shown that Giacomini and Rossi ( 2010 ) tests have undesirable power properties, power that can be low and non‐increasing as the alternative becomes further from the null hypothesis. On the contrary, our statistics are shown to have higher monotonic power, especially the UDmax version. We use their empirical examples to show the practical relevance of the issues raised.     

A simple test of changes in mean in the possible presence of long‐range dependence

We propose a simple testing procedure to test for a change point in the mean of a possibly long‐range dependent time series. Under the null hypothesis, the series is stationary with long‐range dependence and our test statistic converges to a non‐degenerate distribution, whereas under the alternative, the series has a change point in the mean and the test statistic diverges to infinity. We demonstrate the good size and power properties of our test via simulations and illustrate its usefulness by analysing two real data sets.     

Estimation of the location and exponent of the spectral singularity of a long memory process

Abstract.  We consider the estimation of the location of the pole and memory parameter ω ₀ and d of a covariance stationary process with spectral density
We investigate optimal rates of convergence for the estimators of ω ₀ and d , and the consequence that the lack of knowledge of ω ₀ has on the estimation of the memory parameter d . We present estimators which achieve the optimal rates. A small Monte-Carlo study is included to illustrate the finite sample performance of our estimators.     

Consistent estimation of the memory parameter for nonlinear time series

Abstract. For linear processes, semiparametric estimation of the memory parameter, based on the log‐periodogram and local Whittle estimators, has been exhaustively examined and their properties well established. However, except for some specific cases, little is known about the estimation of the memory parameter for nonlinear processes. The purpose of this paper is to provide the general conditions under which the local Whittle estimator of the memory parameter of a stationary process is consistent and to examine its rate of convergence. We show that these conditions are satisfied for linear processes and a wide class of nonlinear models, among others, signal plus noise processes, nonlinear transforms of a Gaussian process ξ_t and exponential generalized autoregressive, conditionally heteroscedastic (EGARCH) models. Special cases where the estimator satisfies the central limit theorem are discussed. The finite‐sample performance of the estimator is investigated in a small Monte Carlo study.     

Testing for structural change of AR model to threshold AR model

The purpose of this article is to develop the likelihood ratio test for the structural change of an AR model to a threshold AR model. It is shown that the log‐likelihood ratio test converges to the maxima of a two‐parameter Gaussian process in distribution. This limiting distribution is novel and we tabulate the critical values. Some simulations are carried out to examine the finite‐sample performance of this test statistic. This article also includes a weak convergence of a two‐parameter marked empirical process, which is of independent interest.